Re: [Xmame] video resources? vector games failing
On Tuesday 03 August 2004 12:02 pm, Michael J. Sherman wrote: Hi all, Whenever I try to start only *vector* games (Asteroids, Battlezone, etc), I get an error from xmame that reads thusly: Using a Visual with a depth of 24bpp. MIT-SHM XV Extensions Available. trying to use... Success. Using Xv Shared Memory Features to speed up Actual bits per pixel = 32... Clearing YUY2 Ok Initializing video effect 2: bitmap depth = 32, display depth = 16 I386 joystick interface initialization... OSD: Warning: No joysticks found disabling joystick support Trimming bitmap from (0,0)-(638,418) to (0,0)-(638,418) info: setting fragsize to 512, numfrags to 5 info: fragsize = 512, numfrags = 5 info: audiodevice /dev/dsp set to 16bit linear mono 22050Hz info: sysdep_dsp: using oss plugin info: sysdep_mixer: using oss plugin X Error of failed request: BadAlloc (insufficient resources for operation) Major opcode of failed request: 141 (XVideo) Minor opcode of failed request: 19 () Serial number of failed request: 48 Current serial number in output stream: 48 Anyone have any ideas why vector games would fail out in this manner, and more importantly, what can be done to fix it? Thanks! -Mike I've been getting similar problems when playing Quantum. Also, I've had Quantum hard-lock X, to the point where only Magic SysRq enables me to gracefully bring down my machine and reboot. pgpfxeIhzJPEM.pgp Description: signature
Re: [Xmame] xml2info - what is this xml2info?
On Wednesday 28 July 2004 10:45 pm, Lawrence Gold wrote: Shyouzou Sugitani wrote: Doug Holland wrote: Didn't build for me. I used makefile.unix, tweaked it a bit to add optimization flags such, but I do not see an xml2info anywhere in there. Me, too. $ make -f makefile.unix xml2info make: `src/xml2info' is up to date. I have to modify src/unix/unix.mak to build xml2info. -xml2info: src/xml2info/xml2info.c +xml2info: src/xml2info/xml2info.o Strange, now I'm having the same problem. I somehow built it before, since it's sitting in my /usr/local/bin directory, but I don't know how. I've applied your change to unix.mak for the next release. Thanks. I finally ended up building it by hand - gcc -o xml2info xml2info.c That worked, though yeah, having it automagically build in the makefile would be good. pgpa83IZ61shf.pgp Description: signature
Re: [Xmame] Frontend question
On Sunday 18 July 2004 12:08 pm, Frank Cox wrote: On Sun, 18 Jul 2004 11:34:01 -0600 Frank Cox [EMAIL PROTECTED] wrote: xmame.x11 -listxml works fine; gives me the list from start to end. gxmame rebuild game list segfaults in the middle of the job. So it appears to be something on the gxmame end of things, not the xmame end, that's blowing up. In the spirit of scientific experimentation, I just tried this: xmame.x11 -listxml | xml2info And it rocked on right through the entire list with no problem. Interesting... Well, after applying the patch, (as found on the xmame web page), gxmame now works, but it doesn't display mameinfo.dat data - just a blank sidebar with my screenshots. Could it be that there are multiple places in gxmame that depend on listinfo? Doug pgpH891gEYAZh.pgp Description: signature
Re: [Xmame] Frontend question
On Thursday 29 July 2004 10:38 am, Doug Holland wrote: Well, after applying the patch, (as found on the xmame web page), gxmame now works, but it doesn't display mameinfo.dat data - just a blank sidebar with my screenshots. Could it be that there are multiple places in gxmame that depend on listinfo? Never mind, operator error on my part, got the mameinfo stuff working again. pgpv8EPALOqOJ.pgp Description: signature
[Xmame] xml2info - what is this xml2info?
I'm in the process of trying to get gxmame to work with xmame 0.84. According to the patch, it's supposed to be piping stuff through xml2info, which I have no idea where to find. How do I find or build this tool? pgpfXhLRUAhtI.pgp Description: signature
Re: [Xmame] Triple buffering
Yep. The effect is called shearing or tearing. The way to get rid of shearing is to refresh the display on the vertical retrace - when the monitor's electron beam finished scanning a frame and returns to the top of the screen. If you double-buffer, and tell the video card to switch buffers during the vertical retrace, there will be no shearing. Unfortunately, IIRC, there is no reliable way to get vertical retrace information from the video card in X. SDL can do this if you're running straight from the console, without X. IIRC, AdvanceMAME running in SDL console mode can sync to vertical retrace and eliminate your problem. So if you have nothing else to do, go bug the X.org developers and demand that they implement a way to sync with vertical retrace in the next version of X. On Thursday 20 May 2004 11:20 pm, Lawrence Gold wrote: Meghazi Fabien wrote: Hi all, I've got visual artefact with xmame when running games with scrollings. It seems to be a double buffering problem. The problem appears using xv but also with dga and sdl (it's even worst with dga and sdl than with xv) I tried mame32 under windows and I experienced the same problem. This kind of problem is more easy to see in shadow dancer by example (the scene between levels : the panel scrolling from right to left then back from left to right, you can easily see that the frame has been displayed, but the first part is the next frame and it has not yet been drawed completely so at the bottom of the picture you can see the previous frame, this results as a splitting line that makes a pixel delay in scrollings) But when I activated the mame32 triple buffering option, the problem disapeared !! I don't find this option under xmame. Is there an equivalent or something ? Is there a plan to include this option in xmame or is it impossible due to X limitations ? I see the same thing here. There used to be a double buffering option for the SDL target, but it's no longer available. I'm not sure what the best solution would be -- either to add the equivalent of MAME32's triple buffering, or to somehow sync the drawing to the monitor's refresh. (The first option sounds like it'd be less troublesome all around.) I'll file your message in my xmame todo folder and make an attempt at a solution eventually. pgpYpI5g0t2MR.pgp Description: signature
Re: [Xmame] mamedb.org
On Sunday 21 March 2004 10:52 pm, Simon Roby wrote: Anybody know what happened to mamedb.org? It was a pretty useful site as a mame.dk-without-the-roms replacement, but now it appears to have completely vanished... There's http://www.klov.com/ - Killer List of Videogames. It's not MAME-specific, but I find lots of good stuff there. pgp0.pgp Description: signature
[Xmame] starwars: the Death Star bug.
I just upgraded to xmame 0.80.1, and I'm still running into the bug where if I run starwars, start the game, select medium or hard difficulty, and play through the tie-fighter stage, the game does not zoom into the Death Star and move to the next stage. After googling and reading the last thread on this topic, it was mentioned that there was some sort of kludge to work around this. Could somebody post the kludge or tell me where I can go to find a fix? Doug pgp0.pgp Description: signature
Re: [Xmame] starwars: the Death Star bug.
On Thursday 18 March 2004 11:54 am, Doug Holland wrote:
I just upgraded to xmame 0.80.1, and I'm still running into the bug where
if I run starwars, start the game, select medium or hard difficulty, and
play through the tie-fighter stage, the game does not zoom into the Death
Star and move to the next stage.
After googling and reading the last thread on this topic, it was mentioned
that there was some sort of kludge to work around this.
Could somebody post the kludge or tell me where I can go to find a fix?
Doug
HAH! Found the fix for the infamous Death Star bug!
Go to src/machine/starwars.c. On or about line 294, you should find the line
ACC=ACC+( ( (long)((A-B)*C) )14 );
Replace it with
ACC += long)((A - B) * C)) 13) + 1) 1;
I found this in an older version of that file (from version 0.69).
The code there read
if (IP15_8 LDC)
{
C = RAMWORD;
/* TODO: this next line is accurate to the schematics,
but doesn't seem to work right */
/* ACC=ACC+( ( (long)((A-B)*C) )14 ); */
/* round the result - this fixes bad trench vectors in
Star Wars */
ACC += long)((A - B) * C)) 13) + 1) 1;
}
Please test this and make sure this hack works.
pgp0.pgp
Description: signature
[Xmame] Autoconfiscate XMAME?
The other day, I downloaded AdvanceMAME, because I wanted synchronization with vertical retrace, which isn't easy in XFree86. The feature I really liked from AdvanceMAME is its build system, using the GNU autotools - automake, autoconf, etc. Everyone here knows what a pain it is to tinker with the Makefile to build xmame. AdvanceMAME's is built with ./configure; make; make install. Is there any chance XMAME could be autoconfiscated to simplify compiling? pgp0.pgp Description: signature
Re: [Xmame] starwars: Workarounds for the death star bug?
On Sat 20 Dec 2003 3:26 pm, Lawrence Gold wrote: Doug Holland wrote: In recent versions of xmame, there is a bug in the starwars driver where if you select medium or hard difficulty, and finish the tie-fighter stage, the death star just sits in the middle of the screen instead of zooming to the next stage. Are there any workarounds for this? I just tried it with 0.77.1 and it zoomed to the next stage without problems. ___ Xmame mailing list [EMAIL PROTECTED] http://toybox.twisted.org.uk/mailman/listinfo/xmame Did you try medium or hard difficulty? I just tried again, using 0.77.1, and got the death star bug again. It is a known bug in the starwars emulation. I just wondered if anyone else found a way to fix it. pgp0.pgp Description: signature
Re: [Xmame] Autoconfiscate XMAME?
On Sat 27 Dec 2003 3:40 pm, Frank Cox wrote: On Sat, 27 Dec 2003 17:38:04 -0500 Simon Roby [EMAIL PROTECTED] wrote: You do have to edit your makefile manually. And type make install too. Sure, it's easier if you only mention less than half of it. And? You'll still have to specify what video output you want (X11, SDL, what-have-you), whether you have joysticks and so forth and so on. So you'll still end up answering questions either way. Forgot to mention: the AdvanceMAME code puts multiple display drivers in one binary, and the configure script automatically detects what's available on your system. So instead of having to make xmame.x11, xmame.sdl, xmame,xgl, etc., you only have to build one binary and then specify which display driver you want on the command line or in an rc file. pgp0.pgp Description: signature
Re: [Xmame] kernel 2.6 performance issues?
On Sat 20 Dec 2003 2:38 pm, mushroomblue wrote: Attached is a diff between my .config and yours. When you upgraded to 2.6, did you use your 2.4 .config file and just run make oldconfig? heck no. I figured that'd be a problem, being that much has changed. they didn't advise it back when 2.2 migrated to 2.4, and I figured it'd be the same with 2.6. :) One performance-related difference I notice right off the bat is that I have kernel preemption enabled and you don't. I wouldn't expect this to have much of an effect on xmame, though. generally, it shouldn't. but the reason it's disabled in that config is because I read that from 2.6.0-test10 on, they've had a weird bug with kernel preemption. so I have one bzImage with it enabled, and one with it disabled. no difference whatsoever, actually. the one thing I'm thinking is it's mostly sound-related, so I'm going to try regressing to the OSS drivers. another thing is that I compile everything I need into a big monolithic kernel (like linus does), and I read on the gentoo boards that alsa has issues when not compiled as a module. Another thing to look into is make sure XFree86 is running at nice priority 0. In kernel 2.4, it was a good idea to bump X's priority to -10, which improves responsiveness. Because of 2.6's new scheduler, that's actually counterproductive, so make sure it's 0. pgp0.pgp Description: signature
Re: [Xmame] GXMame 0.34 if finally out
On Sat 20 Dec 2003 5:07 pm, XulChris wrote: For some reason 1941 (World) is showing up on the Unavailable folder, but I have the ROM and if I click on it, the game runs. Did you try auditing the game? Right click on the game, go to Properties, then click on the Audit tab. GXMame should automatically audit the game right there, tell you if the roms samples are correct, and update its lists. Alternatively, you can go to File, Audit All Games, and get everything up to date. pgp0.pgp Description: signature
Re: [Xmame] Segmentation faults with FXmame
On Fri 19 Dec 2003 7:32 pm, Martin Harte wrote: Hey, I've been having some problems running the 3dfx versions of xmame(xfx and svgafx). Both seg fault on me as soon as they finish loading up a ROM. loading rom 11: ff19-19.bin done 8bit depth PseudoColor X-Visual not available :-( Using 16bpp TrueColor X-Visual Resource info: using Glide version 2.53 info: screen resolution set to 640x480 Using 16bpp video mode info: setting fragsize to 512, numfrags to 5 info: fragsize = 512, numfrags = 6 info: audiodevice /dev/dsp set to 16bit linear mono 22050Hz info: sysdep_dsp: using oss plugin info: sysdep_mixer: using oss plugin Segmentation fault I'm running Slackware 9.0 with kernel 2.4.20. When this happens with xfx the screen goes blank, with svgafx it does not. test3Dfx runs fine on the system so I know that's working. It's not a problem with the 3dfx card itself as I've tried swapping it with another model and the same problem exists. It's also got nothing to do with sound(i tried it with -nosound). It's got nothing to do with X since svgafx seg faults aswell. I've tried a Linux 2.2 kernel and a few different versions of xmame(0.66, 0.70, 0.74 0.77), all to no avail. Non 3dfx verisons of xmame run fine on the system. I'm out of ideas. Any help would be much appreciated. Cheers, Martin. Try building a debug version of xmame (the options to do so are in the Makefile). Also make sure core dumps are enabled on your system. Then when it segfaults, you can do a post-mortem by running gdb, have it crunch through the core file and get a backtrace, which is quite useful for the developers. Or if you have some programming knowledge, you can fix it yourself. Right now, I don't remember all the commands to enable core dumps and operate gdb though - I'd have to RTFM. pgp0.pgp Description: signature
Re: [Xmame] GXMame 0.34 if finally out
On Tue 16 Dec 2003 3:26 pm, Nicos Panayides wrote: On Tue, 2003-12-16 at 10:39, Doug Holland wrote: Looks nice. One problem I've been having is that I can't get options for individual games to override the global defaults. How do I do this? You can thank me for this bug :) In src/io.c at line 1289: - if(g_file_test(filename, G_FILE_TEST_EXISTS)) + if(!g_file_test(filename, G_FILE_TEST_EXISTS)) Does this fix it? Nicos ___ Xmame mailing list [EMAIL PROTECTED] http://toybox.twisted.org.uk/mailman/listinfo/xmame Indeed it does. I was already one step ahead of you and filed a bug report w/ the patch on sourceforge. :D Doug pgp0.pgp Description: signature
Re: [Xmame] Complete set of modelines
On Tue 04 Nov 2003 10:06 am, Simon Roby wrote: Hello all, Does anyone have a (near) complete set of modelines for xmame, preferably including the modes needed to run effects that use 2-3 times the resolution? I have a lot of them (some copied from the provided info in the xmame distribution), but there's plenty of them still missing. Anyone? Regards, Pieter Hulshoff I'm thinking of coding a program that will automatically generate xmame resolutions according to the output of a -listfull and your monitor's refresh rates and max clock. I'm a little busy right now so I didn't have the time to do it lately, but you reminded me of it, so stay tuned... ___ Xmame mailing list [EMAIL PROTECTED] http://toybox.twisted.org.uk/mailman/listinfo/xmame There's an automatic XFree86 Modeline Generator web page at http://xtiming.sourceforge.net/cgi-bin/xtiming.pl I think you'll find that helpful. Meldroc pgp0.pgp Description: signature
Re: [Xmame] Joystick/controller adjustments?
On Mon 01 Sep 2003 4:26 am, Michael Grizzle wrote: Hello, Sorry Im not any help, but I think I have the same or a similar pad. Mine is a Logitech wireless rumblepad. I also have the same problem. Especially playing roadrunner. I cant control the roadrunner at all or its really erratic. I didnt use jscal. I did try using the mode button on the controller, but that only switches control from the directional pad/POV to the left analog stick. I have no problem controlling roadrunner in windows with this stick. I have the same problem when using my older Hammerhead FX gamepad. Its like the analog stick is either wide open or off. I had a hard time setting the contols for the game until i hit the mode button(an led will switch between red and green). Before, when I tried to configure controls to use an analog stick, it would register both x and y axis simultaneously. If you happen across a solution let me know. thanks griz Hmm, I'm thinking a solution may have to be created for Xmame, in the form of options for adjusting axis dead zones, sensitivity, a jitter filter, things like that. As far as I can tell, there's nothing wrong with my controller. When I test it with jstest, it seems to respond fine, and calibration isn't really a problem (although it reports having one more axis and button than it really has.) It's just that the way xmame reads my joystick inputs needs work. Meldroc ___ Xmame mailing list [EMAIL PROTECTED] http://toybox.twisted.org.uk/mailman/listinfo/xmame
[Xmame] Joystick/controller adjustments?
I'm running xmame 0.72, and using a Logitech WingMan RumblePad. I do have the controller working fine - I calibrated it with jscal, tested it, and it seems to work. The problem is that I'm finding the controller to be twitchy. It's very easy to inadvertently input the wrong direction using the analog sticks, leading to many deaths in my games. Is there any way to make adjustments to the controller? I'm talking about tweaking things like sensitivity, dead-zone, jitter filters, etc. After perusing the docs and even the source, it doesn't look like such adjustments exist. Are there any workarounds for tweaking controller response? And may I request that the developers put some joystick adjustments into a future release? Thanks Meldroc ___ Xmame mailing list [EMAIL PROTECTED] http://toybox.twisted.org.uk/mailman/listinfo/xmame
